Alzheimer's disease (AD) is a devastating neurodegenerative disorder. One of the few drugs used to treat AD is memantine, which inhibits N-methyl-d-aspartate receptors (NMDARs), glutamate receptor subtypes found at nearly all vertebrate excitatory synapses. NMDARs are critically involved in many aspects of nervous system function, including learning and memory. Memantine inhibits NMDARs by blocking the ion channel created by NMDARs and occluding current flow through the channel. It is surprising that inhibition of NMDARs slows the cognitive decline associated with AD, since NMDARs are required for memory acquisition. We have proposed that memantine's utility in treating AD derives from the preferential inhibition of specific NMDAR subtypes by memantine. These NMDAR subtypes are thought to be preferentially expressed by inhibitory neurons in the cortex; as a result, memantine reduces cortical inhibition, partially compensating for a reduction in cortical excitation caused by AD. If our hypothesis is correct, then drugs that preferentially inhiit the same NMDARs as memantine, but act in a more selective manner, may be more effective at treating AD than is memantine. In the proposed experiments we first will test the hypothesis that memantine reduces cortical inhibition via subtype-selective inhibition of NMDARs. We then will examine two novel NMDAR subtype selective antagonists that do not work by blocking the NMDAR channel as does memantine, but instead act as negative allosteric modulators (NAMs). We will further characterize the inhibitory properties of the NMDAR NAMs, and will determine whether they are able to reduce cortical inhibition. The experiments make use of a powerful combination of approaches: electrophysiological recordings from recombinant NMDARs, recordings from cortical neurons in mouse brain slices, and multiphoton imaging of Ca2+ concentration in cortical neurons in brain slices. The proposed research first will provide critica tests of an important hypothesis of the mechanism of action of one of the few available AD therapeutics. The results may validate specific NMDAR subtypes as a target for AD treatment; preliminary data strongly support this outcome. Second, the proposed research will test the hypothesis that a new class of compounds may provide a promising new approach for treatment of AD.